key: cord-0008714-0tl1imqo
authors: Couturier, Marc Roger; Bard, Jennifer Dien
title: Direct-from-Specimen Pathogen Identification: Evolution of Syndromic Panels
date: 2019-08-02
journal: Clin Lab Med
DOI: 10.1016/j.cll.2019.05.005
sha: b64c6edaab74205cb71ca9b42fafd277d08ae4e0
doc_id: 8714
cord_uid: 0tl1imqo

This article describes the current state of the art with regards to commercially available syndromic panels for blood stream infections, gastrointestinal pathogen detection, respiratory tract infections, and central nervous system infections, while providing a provocative and speculative look into the future of syndromic panel testing for infectious diseases.

Clinical laboratory testing for pathogens associated with a specific syndrome have historically relied on multiple different analytical approaches to maximize broad pathogen detection. This is especially true for organisms for which culture was not a feasible modality, and instead culture-independent methods such as antigen detection, direct staining, or targeted polymerase chain reaction (PCR) were the aggregate conventional methods. Since the late 2000s, commercial manufacturers of in vitro diagnostic products have invested significant resources into the development of Disclosure Statement: M.R. Couturier receives household income from BioFire diagnostics and is involved in clinical trials activities with Luminex (Verigene). J.D. Bard is a consultant for Bio-Fire Diagnostics and Accelerate Diagnostics and is involved in clinical trials activities with Bio-Fire Diagnostics, Luminex Corporation and DiaSorin Molecular. a ARUP Laboratories, University of Utah, 500 Chipeta Way, Salt Lake City, UT 84108, USA; b Microbiology and Virology Laboratories, Department of Pathology and Laboratory Medicine, Children's Hospital Los Angeles, University of Southern California, 4650 Sunset Boulevard MS#32, Los Angeles, CA 90027, USA multiplex molecular diagnostic assays that are capable of detecting a broad array of pathogens that collectively could cause a single clinical syndrome. This became known as the syndromic panel approach to pathogen testing, and this avenue of test design and development has carried through the recent decade with an evergrowing portfolio of offerings from various in vitro diagnostic manufacturers.

Syndromic testing has carried the field of clinical microbiology to a new frontier of diagnostic capabilities. Although many of the advantages of syndromic testing have been met with equal and opposite disadvantages and challenges, a common opinion is that molecular methodologies for many conventional infections are more accurate, rapid, and convenient than most traditional culture and direct detection techniques previously used by laboratories. But what is next and what should the future of these syndromic panels look like? Where are the gaps and what are the unmet needs? This article describes the current state-of-the-art with regards to commercially available (primary those cleared by the US Food and Drug Administration [FDA]) syndromic panels for respiratory tract infections, gastrointestinal (GI) pathogen detection, blood stream infections, and central nervous system (CNS) infections, while providing a provocative and speculative look into the future of syndromic panel testing for infectious diseases.

Respiratory tract infection panels were the first example of syndromic testing to become commercially available to clinical laboratories approximately 10 or more years ago on a limited number of platforms that have since increased to include multiple different platforms and panel designs. These assays target upper respiratory tract (URT) pathogens from nasopharyngeal swabs or lower respiratory tract pathogens from aspirates or bronchial lavages. The lower respiratory tract panels ( Table 1 ) have only recently been introduced to clinical care, and therefore they are not discussed in detail owing to the unclear nature of their future needs.

URT syndromic testing was a welcomed replacement for insensitive and cumbersome viral cultures and stains as well as imperfect bacterial serology assays. Although PCR for influenza was available to laboratories previously, syndromic panels expanded the detection capacity to include a broad array of common viral targets and select bacteria not traditionally detected by culture. These assays vary in terms of breadth and scope of targets ( Table 2) . The platform ease of use and complexity is also variable. Extreme ends of the spectrum include the BioFire FilmArray Respiratory EZ panel (Bio-Fire Diagnostics, Salt Lake City, UT, USA), which is Clinical Laboratory Improvement Amendments of 1988 waived and can be performed at near point of care, versus the Luminex NxTAG Respiratory Pathogen Panel (Luminex Corporation, Austin, TX USA), which is high complexity and requires molecular laboratory expertise and workspace.

Emerging respiratory viruses such as enterovirus D68 and coronavirus family members have been identified recently in outbreaks affecting multiple countries. 1,2 These emerging and evolving viruses pose the most immediate opportunity for assays to provide more comprehensive detection of respiratory viruses. The FilmArray RP2Plus panel currently includes the Middle East respiratory syndrome coronavirus; however, the clinical performance is not fully understood owing to limited widespread circulation of the virus. This target is suppressed in patients who do not meet clinically compatible presentations. This opens up an interesting future opportunity for flexible testing Direct-from-Specimen Pathogen Identification options related to such pathogens. Because the positive predictive value of a test result will be significantly impaired in the absence of circulating virus, emerging respiratory pathogens (eg, Middle East respiratory syndrome coronavirus, enterovirus D68, severe acute respiratory syndrome) could be included in a syndromic panel, but the end-user could have the option of disabling this target in the absence of documented circulation.

An alternative approach could be that certain targets could be tested but not immediately visible. The results could instead be communicated to the manufacturer through a cloud-based surveillance system. In this way, a sporadic false positive could be easily identified, but a community-level cluster of positives could be identified by the manufacturer and an investigation and communication with the end-user laboratories and treating physicians could ensue. Although this may be an Orwellian, intrusive system, a similar voluntary surveillance system already exists for BioFire users and could have value in future epidemiologic outbreak identification. 3 Rather than continuing to chase outbreaks after the fact, earlier crowd sharing of silent surveillance data could serve the entire community while not sounding an alarm on a sporadic false-positive result. This process would require a discussion of who pays for this extra effort and target detection-manufacturers, public health, or end-users.

Simple, rapid, and near point-of-care testing

The future of URT pathogen testing is ultimately coming to near point-of-care with extremely rapid turn around time. Although microbiology laboratories are hesitant to let go of their testing fiefdom, the logical progression of testing and rapid decision making lies in having answers immediately available and actionable. Examples of technologies that bring single or 2-target respiratory pathogen testing to near point of care for influenza have been embraced in many health care settings (despite imperfect performance characteristics) and have allowed for more efficient use of emergency rooms and decreased costs. 4, 5 Comprehensive URT panels with even shorter turn around time than current assays could allow for improved management of admissions and patient cohorting, while possibly reducing early empiric (and often unnecessary) antibiotic administration.

GI pathogen detection has historically relied upon multiple classic complementary diagnostics modalities (eg, culture, fecal antigen detection, microscopy, single target PCR) to create a comprehensive pseudopanel of targets. A longstanding challenge with respect to GI pathogen detection includes multiple factors:

1. Limited capacities of the local laboratory and lack of usefulness for reference laboratory testing for acute GI illness 2. Physician ordering lapses stemming from a lack of understanding of test methods and detection capabilities 3. Preanalytical and analytical factors that decrease sensitivity of the available methods (eg, delayed transport or preservation of stool, inexperienced technologists, general unfamiliarity with less frequently encountered pathogens)

Syndromic panels have approached these challenges by compartmentalizing pathogens into a single large panel or multiple modular panels grouped by taxonomic relatedness (eg, bacteria, viruses, protozoa).

Small modular panels can detect more common bacteria, viruses, or protozoa. These panels for the most part have targeted the highest prevalence or significance organism at the expense of less comprehensive detection of other pathogens. A panel capable of detecting Salmonella, Shigella, Campylobacter jejuni/coli, and Shiga-like toxin genes of Shiga-toxigenic Escherichia coli was essentially a replacement for conventional stool culture. However, clinical concerns about failing to detect less frequently encountered pathogens such as Vibrio, Yersinia entrocolitica, and enterotoxigenic E coli drove the creation of expanded bacterial panels which could be considered for specific patient populations or patients who initially test negative for the more common bacterial pathogens (Table 3) .

Similarly, standalone parasitic panels primarily targeting the most common protozoal pathogens can be viewed as replacements for stool ova and parasite examinations, with only a rare subset of patients requiring a full ova and parasite examination to rule out helminth infections or less common protozoal pathogens (eg, Cystoisospora, Balantioides). The most common configuration of testing includes Giardia, Cryptosporidium, and Entamoeba histolytica (see Table 3 ).

Modular panels for enteric viral pathogens have not been widely commercialize to date, with only 1 product from BD (Beckton Dickinson, Franklin Lakes, NJ, USA) having clearance from the FDA (see Table 3 ). This may be understandable from a practical standpoint because conventional testing for GI viruses has not been universally embraced as a standard of care, with the exception of rotavirus antigen detection for neonates and PCR for noroviruses in outbreak settings. 6 

Several commercial panels are capable of detecting bacteria, viruses, and protozoa in a single assay ( Table 4 ). These panels vary by specific target in the case of bacteria and viruses; however, a core set of pathogen targets has been included in most commercial assays. These targets are thought to represent the most common enteric pathogens.

Targets of unclear or unestablished significance

To date, GI panels have included definitive pathogens; however, some detect organisms with less definitively established pathogenicity. Such targets include enteroaggregative E coli, enteropathogenic E coli, and Plesiomonas shigelloides. Manufacturers should consider removing or masquing such targets or conducting additional clinical studies to better establish significance in specific test populations before integrating them into the panels. 

A glaring hole in the design of most commercial assays is the notable exclusion of Cyclospora cayetanensis. This pathogen is of significant importance in the Americas and in the recent decade has become a quasiseasonal illness in the United States owing to imported produce serving as a vehicle for multistate and nationwide outbreaks. One study has already demonstrated the value of detecting C cayetanensis directly from stool specimens during an outbreak. 7 Despite this prevalence and clinical/epidemiologic significance, the target has been excluded from multiple panels in favor of E histolytica (see Tables 3 and 4) . Although E histolytica is a virulent protozoal pathogen, its incidence is very low in developed countries. In this author's laboratory, Cyclospora is the most frequently detected protozoal pathogen and E histolytica ranks fifth of the 5 protozoa detected by a syndromic panel. Furthermore, some commercial assays are not able to specifically identify E histolytica from the nonpathogenic Entamoeba dispar. This limitation is essentially the same problem that has plagued ova and parasite examinations for a century. Improving the species-level specificity for E histolytica and inclusion of Cyclospora in future products is imperative.

Testing for C difficile has been an area of clinical diagnosis that has evolved significantly over the past decade. Although the nuances and opinions with regard to this infection are beyond the scope of this work, there are some practical issues that arise as a result of panel tests that include this target. Testing for C difficile in children less than 24 months of age has been traditionally discouraged because children in this age group are often asymptomatically colonized. 8 Furthermore, testing for C difficile in otherwise healthy adult populations without previous use of antibiotics or hospitalization or hospital exposure has not historically been advocated during primary clinical evaluations. 8 Inclusion of this target can increase the risk for further overdiagnosis of C difficile.

Another future area of consideration for aiding in the interpretation of colonization versus infectious state could include the integration of mucosal inflammatory marker expression, for example, proinflammatory cytokines. Markers of acute inflammation could serve as an adjunct metric to aid in interpreting the detection of potential colonizers like C difficile or organisms capable of shedding for week or month after convalescence (eg, Salmonella and norovirus). This would be an area still in need of significant research because molecular profiling for inflammatory bowel disease and other similar conditions is not yet a standard of care.

Demographic or region-specific assays (or convenient end-user customization) should be an area of future consideration for manufacturers. Future GI syndromic testing must be customizable to fit the needs of the individual laboratory or region while not overtesting for targets that have low to no prevalence in the region or demographic. For example, although including Cyclospora in a syndromic panel may be very relevant for the Americas and tropical/subtropical countries, it may be a completely unnecessary target in other regions (eg, Canada, Northern and Eastern Europe). Although rare infections may be encountered in these regions, it may not be cost effective to test all specimens for this target and in the absence of true infections, the positive predictive value of this target will be poor. Considering demographic-driven testing, a community hospital with primarily uncomplicated patients would have a different pathogen list to consider for clinical care than a large academic medical center that supports a transplant center, human immunodeficiency virus clinic, and travel/tropical medicine clinic.

A flexible end-user design already exists in the Verigene (Luminex Corp., Austin, TX, USA) RP Flex assay.

One challenge that is unmet by the current GI syndromic testing is the detection of antimicrobial resistance to primary empiric antibiotics. This is a daunting technical challenge in a specimen such as stool that contains copious genera of commensal microbiota; however, even a few targets for primary resistance could be invaluable to clinical care for some enteric bacterial pathogens. Organisms such as Shigella, Salmonella, and Campylobacter are currently cultured after syndromic testing in efforts to maintain public health outbreak tracking, but also to provide primary antimicrobial susceptibility testing for selected drug classes in cases of severe infection. 9,10 These organisms often are not culturable after primary syndromic testing, so having markers for macrolide, quinolone, and/or tetracycline resistance could be extremely helpful for treating clinicians and epidemiologic reporting. 11 Clearly this is a technically challenging request and one that may be more idealistic than realistic given current technologies, but should nonetheless be a future target goal for testing.

Future syndromic testing for GI pathogens should consider 1. Customizable or regional/demographic-driven targets 2. Inclusion of only clinically relevant targets or those with well-established clinical significance 3. Inclusion of inflammatory markers to aid in identifying potential colonizers versus active infections 4. Inclusion of antimicrobial resistance determinants for outbreak associated bacterial pathogens

Expeditious identification of bloodstream pathogens through the use of molecular syndromic panels has dramatically altered the standard of care in many laboratories. There are currently 6 FDA-cleared multiplexed assays directly from positive blood cultures ( Table 5 )-4 bacterial panels coupled with resistance determinants, 1 fungal panel, and 1 panel that generates minimum inhibitory concentration in lieu of resistance markers. There are currently 2 direct from whole blood panels.

Expansion of existing panels to include additional pathogens and more comprehensive resistance determinants is ongoing. The ePlex BCID-GP panel is currently the most comprehensive panel for gram-positive organisms and also includes pangram-negative and pan-Candida targets as a safeguard for Gram stain interpretation (see Table 5 ). Additional gram-positive organisms that may be considered are Streptococcus mitis/oralis, an important bloodstream infection (BSI) agent in patients with underlying hematologic and oncologic diseases, and Corynebacterium jeikeium. The inclusion of targets for Mycobacterium species, particularly the rapid-growing nontuberculosis Mycobacterium, would also be beneficial for certain patient demographics. A relatively underappreciated pathogen group are anaerobes, accounting for approximately 20% of BSIs. 12 An up and coming gram-negative panel (BCID-GN) targets numerous obligate anaerobes, including Fusobacterium necrophorum, F nucleatum, and Bacteriodes fragilis. Likewise, the next generation of the FilmArray BCID (BCID2) will include B fragilis. Inclusion of additional obligate anaerobes (eg, Clostridium spp.) would further expand the breadth of coverage offered in these syndromic panels. The inclusion of select agent targets such as Brucella spp, Franciscella spp., and Burkholderia pseudomallei may prevent unnecessary laboratory exposure. This request may be onerous for manufacturers to fulfill owing to restrictions associated with select agent testing. Last, the development of a fungal panel that includes comprehensive list of molds would significantly improve time to identification. The ePlex BCID-FP panel offers detection of Fusarium spp. (see Table 5 ), but some additional relevant targets to consider are Scedosporium apiospermum and Lomentospora prolificans.

There is a crucial unmet need to bypass the primary incubation step entirely. Exciting new technologies are currently in development, and an excellent summary of these technologies are provided by Sinha and colleagues. 13 The first innovative step to this holy grail approach is the release of 2 FDA-cleared, whole blood assays (see Table 5 ) that target 5 bacteria and Candida species. This is a significant milestone in BSI diagnostics and provides insight for future panels. For starters, whole blood syndromic testing is an adjunct to conventional blood cultures and manufacturers must consider the pediatric population and associated blood volume limitations. A requirement of less than 1 mL of blood in children would be optimal. Second, the targets incorporated must be analogous to the aforementioned direct from positive blood culture panels. Preferably, a compendious list of bacterial, viral, and fungal pathogens should be offered, without compromising the sensitivity of the test. The ability to detect polymicrobial infections despite low pathogen load is paramount for appropriate antimicrobial coverage. Importantly, the breadth of targets and high sensitivity will allow for high confidence in a negative result to potentially decrease antimicrobial use. An overarching goal is to be as target agnostic as possible while maintaining rapidity.

There are a number of caveats to whole blood syndromic testing that needs to be rectified before widespread adoption. First, the increase risk for contamination exists, particularly if extensive manual manipulation is required. Emerging technologies are attempting to alleviate this risk by innovative approaches, including the use of small reaction mixture volumes to reduce the number of contaminating DNA. 14 Second, current BSI syndromic panels pricing is not conducive to testing on all patients with corresponding blood cultures. Manufacturers must be aware of the budgetary limitations instilled on laboratories that may prohibit adoption. The development of a diagnostic algorithm that integrates host response as a predicate of whole blood syndromic testing may be a potential strategy to distinguish low risk patients from septic patients. The SeptiCyte (Immunexpress Inc., Seattle, WA), a recently FDA-cleared quantitative reverse transcriptase PCR based assay, measures levels of biomarkers (CEACAM4, LAMP1, PLA2G7, and PLAC8) to distinguish between sepsis and noninfectious process in adult patients only. 15 A pilot study of 70 pediatric patients also demonstrated promising results. 16 A noteworthy limitation is the required 2.5 mL for testing and future development of host response assays should also be cognizant of blood volume constraints.

Inarguably, the impact of timely susceptibility results is profound because it allows for antimicrobial optimization. New generations of BSI syndromic panels currently in development have expanded the list of resistance targets of epidemiologic and therapeutic importance including, mecC and mcr-1 genes. However, targets to detect extended-spectrum beta-lactamase production remains limited to the bla CTX-M gene and would benefit from additional genes, including bla SHV and bla TEM .

In contrast with genotypic resistance detection offered by the majority of panels, there is a paucity of rapid phenotypic susceptibility platforms. Although both testing modalities have clinical usefulness, only panels that provide simultaneous identification and phenotypic susceptibility results can function as a standalone test. Expansion of the identification capabilities of the existing PhenoTest BC (Accelerate Diagnostics, Tucson, AZ, USA) would be beneficial, because the gram-positive targets are currently limited (see Table 5 ). Manufacturers need to also be cognizant of prioritizing the release of susceptibility testing options for novel antimicrobial agents and to accommodate clinical breakpoints changes.

Finally, the amalgamation of direct whole blood syndromic panel with phenotypic susceptibility testing is an ambitious request that may be realistically fulfilled in the near few years.

Laboratory diagnostic tests are compulsory to confirm CNS infection. The diagnostic accuracy of conventional microbiological approaches, such as Gram stain and culture, is hampered by low sensitivity and/or slow turnaround time. New and emerging diagnostic approaches to aid in the diagnosis of meningitis and encephalitis may address the limitations of current laboratory practices.

There are currently only 3 FDA-cleared tests for the detection of pathogens directly from cerebrospinal fluid (CSF) specimens ( Table 6 ). The first 2 are limited-target, Other key points about current syndromic testing for CNS infections include:

1. Increased pathogen detection compared with conventional laboratory approaches 2. Associated risks of contamination necessitate the need for strict adherence to molecular testing policies 3. Low viral load in CSF for certain viruses (eg, herpes simplex virus) may warrant additional testing from alternate sources (eg, blood) 4. Molecular testing may be suboptimal for certain pathogens (eg, cryptococcal antigen testing continues to be the diagnostic standard) 5. Detection of herpes viruses could represent either latent or active infection. Chromosomal integration is also a possibility in patients positive for human herpes virus-6 (inherited chromosomally integrated human herpes virus-6). 6. Corroboration of findings with clinical picture is paramount to ensure the most appropriate diagnosis and patient management.

Syndromic testing for CNS infections remains in its infancy. Continuous innovation and development are fundamental to the diagnostic advancement of CNS infections. A quick fix that current and future manufacturers may consider would be to decrease the required CSF volume to accommodate the copious numbers of tests often ordered.

It may be valuable to offer customizable panels for acute versus chronic CNS infections. Acute meningitis is often caused by viruses and bacteria included in the FilmArray Meningitis/Encephalitis panel. In contrast, chronic meningitis can result from infection with Mycobacterium tuberculosis and fungal pathogens including, Aspergillus, Histoplasma, Blastomyces, and Coccidiodes. 17 The inclusion of additional, uncultivable pathogens, may add further value to the panel (eg, Bartonella spp. and Treponema pallidum). This demographic consideration can be expanded geographically to include pathogens that may be applicable to certain regions.

An entire niche of CNS infections remains untouched. Cumulative incidence rates of shunt infections range from 10% to 22% per patient with 90% of infections occurring within 30 days of surgery. 18 No commercial molecular tests exist for the diagnosis of shunt infections and microbiology culture remains the primary approach. The microbiological demographic of shunt infections is attributed to skin flora and differ significantly from patients with acute meningitis and encephalitis. As such, the FilmArray Meningitis/ Encephalitis panel is not appropriate for patients with indwelling devices and also not approved by the FDA for such testing. Laboratory-developed molecular tests that detect Staphylococcus aureus and Cutibacterium acnes from shunt specimens have yielded promising data, including increased detection in patients with prior antimicrobial exposure. 19 We envision future assays that target common shunt pathogens including, S aureus, C acnes, coagulase-negative staphylococci, gram-negative organisms, and Candida spp. A foreseen limitation would be the risk of skin flora contamination, which may impede the inclusion of coagulase-negative staphylococci target.

Arbovirus infections are important causes of neuroinvasive diseases associated with endemic transmission or travel-related mosquito exposures. Molecular testing of CSF specimen may be inferior in cases of arbovirus infections because the window for viral detection is mere days from symptom onset compared with the measurement of IgM and IgG antibody responses, which remain elevated for weeks. 20 Thus, diagnosis is made primarily by serology. However, the superiority of serologic testing may be hampered in immunosuppressed patients, stressing the need for enhanced diagnostic tools. Recent studies revealed promising results for detection of West Nile virus from whole blood and urine. 21, 22 A retrospective study of 38 patients with West Nile virus confirmed superiority of molecular detection of West Nile virus in whole blood compared with CSF with a sensitivity of 86.8% and 16.6%, respectively. 21 Yet molecular testing of CSF specimens may still be valuable. 23 Potential syndromic panel for routine detection of arboviruses may integrate molecular testing from whole blood and CSF alongside serologic testing to improve diagnostic yield.

Owing to an active innate immune system in the CNS that rapidly responds to alterations in CNS homeostasis, the levels of biomarkers and cytokines released into the CSF may be diagnostic discriminators. Future syndromic panels may incorporate molecular target detection alongside markers of host responses. Studies on the role of biomarkers in predicting CNS infections and to differentiate bacterial from viral meningitis is ongoing. CSF lactate may be a promising marker to rule out bacterial meningitis with high sensitivity (93%-95%) and specificity (95%-96%), 24 albeit a significant decrease in sensitivity is noted with antibiotic exposure. 24 A recent proof-of-concept study using the Luminex FlexMPA 3D technology determined that quantification of cytokine levels in the CSF can rule in or out an infectious process. Specifically, high levels of IP-10/CXCL10 were present in CNS infections and MDC/CCL22 levels were significantly higher in nonviral infections. Deciphering between oncologic process and autoimmune encephalitis may also be possible, as indicated by high levels of GRO/CXCL1, IL-7, and IL-8 in gliomas. 25 A potential steward of diagnostic testing would include screening for infectious processes by cytokine profiling followed by subsequent testing with syndromic panel. This method could potentially provide decipher between latent versus active viral infection or cases of chromosomally integrated human herpes virus-6 (iciHHV-6). Again, the integration of this algorithmic approach must maintain STAT testing capabilities to maximize the benefit of expeditious testing currently offered by syndromic panels. Extensive research is required to determine true clinical performance and usefulness of such tests.

Inclusions of potential resistance markers and/or direct from specimen phenotypic susceptibility may be beneficial for some CNS pathogens including, Streptococcus pneumoniae, Haemophilus influenzae, and E coli K1. One example is to establish H influenzae antimicrobial profile as beta-lactamase positive, beta-lactamase negative ampicillin susceptible, and beta-lactamase positive ampicillin resistant by the detection of such genes as TEM-1, ROB-1, and PBP3. 26 In line with potential assays for shunt infections, inclusion of mecA and mecC gene for detection of methicillin resistance in S aureus is beneficial. An idealistic request is to develop direct from positive CSF specimen phenotypic susceptibility testing. Future development may consider either a standalone identification and susceptibility testing platform or in collaboration with existing identification platforms to consolidate the susceptibility portion.

There are numerous areas of opportunity to advance the role of syndromic panels in clinical care for respiratory, GI, blood stream infections, and CNS. Based on the primary points of this article, the future of syndromic testing should include:

1. Refining existing targets and inclusion of emerging pathogen targets 2. Expansion of specimen types and volumes required for testing 3. Integration of host response markers 4. Inclusion of additional antimicrobial resistance determinants and phenotypic susceptibility 5. Simple, fast, near point-of-care platforms

First confirmed cases of Middle East respiratory syndrome coronavirus (MERS-CoV) infection in the United States, updated information on the epidemiology of MERS-CoV infection, and guidance for the public, clinicians, and public health authorities

Severe respiratory illness associated with enterovirus D68 -Missouri and Illinois

Automated real-time collection of pathogen-specific diagnostic data: syndromic infectious disease epidemiology

Clinical impact of rapid point-ofcare PCR influenza testing in an urgent care setting: a single-center study

Economic and operational impact of an improved pathway using rapid molecular diagnostic testing for patients with influenza-like illness in a German emergency department

Laboratory diagnosis of noroviruses: present and future

Implications of culture-independent panel-based detection of Cyclospora cayetanensis

2017 infectious diseases Society of America clinical practice guidelines for the diagnosis and management of infectious diarrhea

A challenge and an opportunity to improve patient management and public health surveillance for food-borne infections through culture-independent diagnostics

Clinical microbiology laboratories' adoption of culture-independent diagnostic tests is a threat to foodborne-disease surveillance in the United States

Impact of culture-independent diagnostic testing on recovery of enteric bacterial infections

Anaerobic bacteremia

Emerging technologies for molecular diagnosis of sepsis

Universal digital high-resolution melt: a novel approach to broad-based profiling of heterogeneous biological samples

A molecular host response assay to discriminate between sepsis and infection-negative systemic inflammation in critically ill patients: discovery and validation in independent cohorts

Diagnostic accuracy of a host gene expression signature that discriminates clinical severe sepsis syndrome and infection-negative systemic inflammation among critically ill children

Acute versus subacute community-acquired meningitis: analysis of 611 patients

Ventricular shunt infections: immunopathogenesis and clinical management

Polymerase chain reaction for the rapid detection of cerebrospinal fluid shunt or ventriculostomy infections

Virus and antibody dynamics in acute west Nile virus infection

Superiority of West Nile virus RNA detection in whole blood for diagnosis of acute infection

Excretion of West Nile virus in urine during acute infection

Molecular characterization of viruses associated with encephalitis in Sao Paulo

Diagnostic accuracy of cerebrospinal fluid lactate for differentiating bacterial meningitis from aseptic meningitis: a meta-analysis

Potential role of CSF cytokine profiles in discriminating infectious from non-infectious CNS disorders

Antimicrobial resistance in Haemophilus influenzae

Direct-from-Specimen Pathogen Identification